BOOK: Jacopo Zabarella, "On Methods"

Jacopo Zabarella: On Methods. Vol. 1: Books I–III. Edited and translated by John P. McCaskey. The I Tatti Renaissance Library, Vol. 58. Harvard University Press, 2013. 9780674724792. xxvi + 323 pp.

Jacopo Zabarella: On Methods. Vol. 2: Books III–IV. On Regressus. Edited and translated by John P. McCaskey. The I Tatti Renaissance Library, Vol. 59. Harvard University Press, 2013. 9780674724808. vi + 470 pp.

Zabarella was a 16th-century philosopher who, according to the introduction in this book, was mostly interested in logic and closely followed Aristotle's ideas. His books, too, were to a considerable extent written as commentaries on Aristotle's works. In the present book, On Methods, Zabarella often refers to Aristotle's Posterior Analytics and presents his own ideas as either explanations of Aristotle's work or as polemics against other philosophers who understood Aristotle differently than Zabarella did. Much like with previous philosophy books in the I Tatti Renaissance Library, I'm not really the right reader for this book; I know pretty much nothing about logic as it is understood by Aristotle and similar people, and I haven't read any of his books about logic. From Zabarella's work it is clear that he was part of a lively academic culture in which logicians argued back and forth about various details of their field, and his book is likewise a part of these arguments; but I know too little of the background to be able to really appreciate the details of what they argued about, or to be able to see why these things were so important to them.

Still, I found this book not entirely uninteresting when read in small increments, and it gave me at least a rough idea of some of the things that philosophers who studied logic were interested in. Zabarella starts by introducing method in the broader sense as “an instrumental habit of the understanding by which we are helped in gaining knowledge of things” (1.2.3). He then divides it into order (how to arrange the parts of a discipline when teaching it so that it can be learned as easily and optimally as possible; 1.11.2) and method in the narrower sense (how to get from a known thing to something not yet known in the pursuit of new knowledge). Thus, the key difference is that method (in this narrower sense) involves some sort of inference, where one thing necessarily follows from another (3.2); order, on the other hand, isn't argumentation and one thing doesn't necessarly follow from another, though one thing may direct us towards things that will be explained later in the order (3.1).

I was rather surprised by this idea that the order of teaching a discipline (which for Zabarella could be either a scientific field or a more practical “art” (i.e. skill), such as architecture or medicine) should be investigated in such a general way by philosophers. I naively thought that, if someone is knowledgeable about some field and is trying to write a textbook about it (or set up a course of lectures), he will have no particular difficulty in arranging the various parts of the field into a suitable order; surely you just have to follow common sense and make sure that in your order, if some topic is needed to understand or explain some other topic, then the former has to appear earlier in your order than the latter. And some of the details of your order might be better determined not by arguing philosophically in advance, but simply by some experimentation: try teaching in various orders and see which of them is easier for the students to learn. Why would it need to be any more complicated than that?

But philosophers seem to have taken a great deal of pleasure in making things more complicated than that. Zabarella mentions the traditional division of orders into three kinds, depending on where you start (1.5, 2.2): compositive or synthetical (from causes to effects), resolutive or analytical (from effects to acuses), and definitive (from definitions). He objects to the idea that where you start should be the key thing that determines what kind of order you have (2.3), and he particularly objects to the definitive order, saying that it doesn't really exist at all: a definition just names something, it doesn't really explain anything; if you remove the definitions from the start of your treatment, the rest of it will then be seen to be either in resolutive or compositive order, so that was really the true order of your treatment all along (2.4).

So his view is that there are only two kinds of order (2.6): compositive (used to teach contemplative sciences, i.e. those where we seek knowledge for its own sake; the order begins from beginning-principles) and resolutive (used to teach practical arts; the order begins from the end (purpose) that we want to achieve with that art; e.g. for medicine, that's healing the patient; for architecture, it's building a house; etc.). Thus, he says that for any given science of art, only one or the other of these two orders is suitable (but he does then admit that for some smaller sub-areas of that discipline, the opposite order might be better; 2.20). He gives numerous examples from the works of various classical authors, especially Aristotle and Avicenna (2.10–14).

In Book 3, Zabarella focuses on method. As mentioned above, method necessarily involves some sort of inference; thus it is very similar to syllogism, the difference being that syllogism infers from what is posited, while method infers from what is known. In other words, syllogism can be about something hypothetical or it can start from something that isn't actually known, while method always works from something known to something unknown (3.3).

According to Zabarella, there are two knds of method (3.4): demonstrative (from cause to effect) and resolutive (from effect to cause). He argues vigorously against other authors who mentioned the existence of two other methods, namely divisive and definitive. Division (e.g. of a genus into its species), he says, is more like an order than like a method; it doesn't generate new knowledge, but it can be useful for arranging the parts of an already known science (3.10). Similarly, definition is not a method at all, as it doesn't lead to new knowledge about anything — it just clarifies the signification of a word (3.15).

Of the two kinds of method, the demonstrative one is more important, while resolutive is more of an auxiliary thing. With resolutive method, you discover the beginning-principles of a thing, and once you have them you use the demonstrative method to learn about things from their causes (3.18). He often emphasizes that to really know a thing, you must know not only what it is, but “what it is on account of” (i.e. its causes), and this is what demonstrative method leads to.

Resolutive method can be further divided into demonstration (= inference) from effects (which leads to the discovery of causes) and induction (from particulars to universals); 3.19.

Book 4 is mostly an assortment of Zabarella's opinions about various debates regarding Aristotle's Posterior Analytics; for me it was the least interesting part of the whole work. There seem to have been a surprising number of arguments about what exactly Aristotle's book is about — it strikes me as a bit disturbing that such a foundational work of some field should turn out to be so unclear and have so many different interpretations, but perhaps in philosophy this is actually a plus rather than a minus. Still, there are a few interesting ideas about the relationship between definition and demonstration: demonstration is a logical instrument (a method), definition is the end (purpose) of that instrument. Demonstration has discursive movement (inference), definition doesn't (4.16); the purpose of demonstration is to obtain a definition of the thing being investigated (4.17).

On Regressus

This is a separate short work included at the end of volume 2 and I found it fairly interesting as I had never heard of the regressus before. According to Zabarella, this is a logical procedure somewhat similar to a circular argument. In a circular argument, you first demonstrate a conclusion from the premises (major and minor) and then in the next step invert this and demonstrate both premises from the conclusion; both of these steps use the demonstrative method, going from distinct knowledge of one thing to distinct knowledge of another. The circular argument, of course, famously doesn't make any sense and shouldn't be used.

In the regressus, you start with what Zabarella calls “confused” or imperfect knowledge of some effect, and from it you use the resolutive method to obtain confused knowledge of its causes. There is then an intermediate step to get to “distinct” (or perfect) knowledge of the causes, and then the second step of the regressus where you use the demonstrative method to get from distinct knowledge of the causes to distinct knowledge of the effects. Thus, the net result is that you started with confused knowledge of the effect and ended with distinct knowledge of it. This is a legitimate move and you have gained some knowledge in the process. Another difference compared with the circular argument is that there the second step demonstrated both premises from the conclusion, while in the regressus the second step only demonstrates the minor premise.

Zabarella illustrates these things with a couple of examples from Aristotle, of which I particularly liked the one about smoke and fire. Suppose we see some smoke (that is the effect). We have confused knowledge of it — we know that it is there, but we don't know what it is “on account of”, i.e. what caused it. But we know that where there is smoke, there must be fire (major premise). Together with the fact that there is smoke (minor premise), we obtain the conclusion that there must be fire. This was the first step of the regressus. In the second step, now that we know that there is fire, and we also know that fire causes smoke, we can conclude that the smoke we're looking at must have been caused by the fire. Thus we now have distinct knowledge of the smoke — we know not only that it exists, but what has caused it.

Nevertheless, fun though this may be, I don't have the impression that we have really gained much knowledge in the process. It all seems rather trivial and relies heavily on knowing about the connection between smoke and fire — and it's connections like this that really give us an understanding of nature; and they are in general not that easy to figure out (except in some obvious cases like smoke and fire), and I don't have the impression that the regressus could help us much with that part.

*

As I said earlier, in a way some of the things in this book are somewhat interesting, but I can't really say that I understood what the point of any of this is supposed to be. Would knowing what Zabarella says here about orders really help anyone who is trying to teach some branch of science or some practical art? Would knowing what he says about methods really help any scientist who is trying to discover new knowledge in his field of learning? I can hardly imagine this being the case. I doubt that very many scientists nowadays spend much time thinking about methods in the broad, generic sense that we encounter in this book. Zabarella often gives examples from ancient medical works, from Galen and the like, which are of course invariably grotesquely wrong by our present-day standards — so when he confidently talks about the perfect knowledge of things that we can supposedly arrive at by this or that method, I have a hard time taking him very seriously.

Just like in that old saying about there being no royal road to geometry, I suspect that there is also no royal road to science, no method that would easily and reliably lead to new knowledge; so spending too much time thinking about methods in the generic sense that is done here by Aristotle, Zabarella and other such people strikes me as, well, not terribly useful. But I shouldn't criticize them for it; no doubt the things they did were necessary in their time, forming a phase that human thought had to go through until the conditions were ripe for the development of the more modern approach to science, which indeed was getting started not long after Zabarella's time (though his work seems to be looking more backwards than forwards; cf. the introduction, vol. 1, p. viii).